Biosystems Engineering最新文献_第2页

A novel unsupervised algorithm for pig anomaly detection using video frame prediction 一种基于视频帧预测的猪类异常检测新算法

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2026-01-14 DOI: 10.1016/j.biosystemseng.2026.104383

Zezhong Chen , Qiumei Yang , Deqin Xiao , Jiyan Wu , Manting Wu , Qiwei Hong

Detecting abnormal behaviours in pigs is crucial for enhancing pig welfare. Current research on pig anomaly detection primarily relies on supervised learning methods, facing challenges such as limited generalisability, the complexity of sample annotation, and the inability to cover all abnormal scenarios. To tackle these challenges, an unsupervised video anomaly detection algorithm for pigs based on future frame prediction (PigVADNet) is proposed. PigVADNet is developed to address the unpredictability of abnormalities in pig production. It accurately predicts normal pig behaviours by learning from video frames depicting normal pig behaviours. When the video frames capture abnormal behaviours, there is a significant increase in prediction error, which enables the detection of anomalies in pigs. The model employs a generative adversarial network architecture consisting of a pig image generator, discriminator, and motion information extraction module. The generator leverages a U-Net with an SSPCAB (Spatial and Spectral Pyramid Channel Attention Block) module to predict future frames. The discriminator improves the generator via adversarial learning, ensuring realistic frame generation. The motion extraction module, combined with appearance and motion consistency losses, enhances the prediction of appearance and motion. Finally, the difference between predicted and real frames is evaluated to detect pig abnormalities. The model achieved an AUC (Area Under the ROC Curve) of 95.1 % on the Pig Video Anomaly Detection Dataset. The experimental results demonstrate that this approach can automatically detect pig anomalies without relying on labelled data. It enables timely interventions to enhance pig welfare and optimise production efficiency.

检测猪的异常行为对提高猪的福利至关重要。目前对猪异常检测的研究主要依赖于监督学习方法，面临着通用性有限、样本标注复杂、无法覆盖所有异常场景等挑战。为了解决这些问题，提出了一种基于未来帧预测的猪视频无监督异常检测算法（PigVADNet）。PigVADNet的开发是为了解决生猪生产异常的不可预测性。它通过学习描述猪正常行为的视频帧来准确预测猪的正常行为。当视频帧捕捉到异常行为时，预测误差显著增加，从而能够检测到猪的异常情况。该模型采用由猪图像生成器、鉴别器和运动信息提取模块组成的生成对抗网络架构。该发生器利用带有SSPCAB（空间和光谱金字塔通道注意块）模块的U-Net来预测未来的帧。鉴别器通过对抗学习对生成器进行改进，保证了生成帧的真实感。运动提取模块结合外观和运动一致性损失，增强了对外观和运动的预测。最后，评估预测帧和真实帧之间的差异，以检测猪的异常。该模型在猪视频异常检测数据集上实现了95.1%的AUC （ROC曲线下面积）。实验结果表明，该方法可以在不依赖标记数据的情况下自动检测猪的异常。它能够及时干预，提高猪的福利和优化生产效率。

{"title":"A novel unsupervised algorithm for pig anomaly detection using video frame prediction","authors":"Zezhong Chen , Qiumei Yang , Deqin Xiao , Jiyan Wu , Manting Wu , Qiwei Hong","doi":"10.1016/j.biosystemseng.2026.104383","DOIUrl":"10.1016/j.biosystemseng.2026.104383","url":null,"abstract":"<div><div>Detecting abnormal behaviours in pigs is crucial for enhancing pig welfare. Current research on pig anomaly detection primarily relies on supervised learning methods, facing challenges such as limited generalisability, the complexity of sample annotation, and the inability to cover all abnormal scenarios. To tackle these challenges, an unsupervised video anomaly detection algorithm for pigs based on future frame prediction (PigVADNet) is proposed. PigVADNet is developed to address the unpredictability of abnormalities in pig production. It accurately predicts normal pig behaviours by learning from video frames depicting normal pig behaviours. When the video frames capture abnormal behaviours, there is a significant increase in prediction error, which enables the detection of anomalies in pigs. The model employs a generative adversarial network architecture consisting of a pig image generator, discriminator, and motion information extraction module. The generator leverages a U-Net with an SSPCAB (Spatial and Spectral Pyramid Channel Attention Block) module to predict future frames. The discriminator improves the generator via adversarial learning, ensuring realistic frame generation. The motion extraction module, combined with appearance and motion consistency losses, enhances the prediction of appearance and motion. Finally, the difference between predicted and real frames is evaluated to detect pig abnormalities. The model achieved an AUC (Area Under the ROC Curve) of 95.1 % on the Pig Video Anomaly Detection Dataset. The experimental results demonstrate that this approach can automatically detect pig anomalies without relying on labelled data. It enables timely interventions to enhance pig welfare and optimise production efficiency.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104383"},"PeriodicalIF":5.3,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The association between transition management and modelled milk yield in multiparous dairy cows 多产奶牛过渡管理与模拟产奶量之间的关系

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2026-01-10 DOI: 10.1016/j.biosystemseng.2025.104382

M. Salamone , M. Hostens , E. Canniere , T. Goossens , V.W.M. van Beest , T. van Gasteren , G. Opsomer , B. Aernouts , I. Adriaens

The transition period remains one of the major challenges for dairy cows. Although farms have adopted automated technological means to assist them in breeding and feeding management, automated tools to tailor the transition management are scarce. Previous work has developed a methodology to calculate the milk yield residuals in the transition period (MRT) and investigated the potential of this feature to monitor multiparous dairy cows in early lactation. The MRT is obtained by subtracting a prediction of the expected production from the realised production at the first test day of the lactation. The current study investigates the associations of MRT compiled at farm level with farm management practices. Therefore, transition management practices were gathered through a survey on 45 Flemish and Dutch farms. After data aggregation and cleaning, data from 33 farms were retained and analysed using a partial least squares regression model. This model has the mean MRT of the individual farms as dependent variable. The independent variables included the survey answers combined with general farm production metrics. In cross-validation, the final model had an R² of 0.69. The regression coefficients of this model mainly revealed that the mean MRT was positively associated with the herd mean 305-day milk yield and negatively associated with the feeding of grassland products in the dry period. While this study reveals valuable insights into the relationship between MRT and farm management practices, limitations such as sample size and unbalanced data distributions warrant further investigation with larger datasets and longitudinal approaches.

过渡时期仍然是奶牛面临的主要挑战之一。虽然养殖场已经采用自动化技术手段来辅助养殖和饲养管理，但用于定制过渡管理的自动化工具却很少。以前的工作已经开发出一种方法来计算过渡期的产奶量剩余（MRT），并研究了这一特征在哺乳早期监测多产奶牛的潜力。MRT是通过在哺乳的第一个测试日从实现的产量中减去预期产量的预测得到的。目前的研究调查了在农场层面编制的MRT与农场管理实践的关系。因此，通过对45个佛兰德和荷兰农场的调查，收集了过渡管理做法。在数据汇总和清理后，保留了33个农场的数据，并使用偏最小二乘回归模型进行了分析。该模型以个体农场的平均MRT作为因变量。自变量包括调查答案与一般农业生产指标相结合。在交叉验证中，最终模型的R2为0.69。该模型的回归系数主要表明，平均MRT与牧群平均305日产奶量呈正相关，与旱季牧草采食量呈负相关。虽然这项研究揭示了MRT与农场管理实践之间关系的有价值的见解，但样本量和不平衡数据分布等限制需要进一步研究更大的数据集和纵向方法。

{"title":"The association between transition management and modelled milk yield in multiparous dairy cows","authors":"M. Salamone , M. Hostens , E. Canniere , T. Goossens , V.W.M. van Beest , T. van Gasteren , G. Opsomer , B. Aernouts , I. Adriaens","doi":"10.1016/j.biosystemseng.2025.104382","DOIUrl":"10.1016/j.biosystemseng.2025.104382","url":null,"abstract":"<div><div>The transition period remains one of the major challenges for dairy cows. Although farms have adopted automated technological means to assist them in breeding and feeding management, automated tools to tailor the transition management are scarce. Previous work has developed a methodology to calculate the milk yield residuals in the transition period (MRT) and investigated the potential of this feature to monitor multiparous dairy cows in early lactation. The MRT is obtained by subtracting a prediction of the expected production from the realised production at the first test day of the lactation. The current study investigates the associations of MRT compiled at farm level with farm management practices. Therefore, transition management practices were gathered through a survey on 45 Flemish and Dutch farms. After data aggregation and cleaning, data from 33 farms were retained and analysed using a partial least squares regression model. This model has the mean MRT of the individual farms as dependent variable. The independent variables included the survey answers combined with general farm production metrics. In cross-validation, the final model had an R<sup>2</sup> of 0.69. The regression coefficients of this model mainly revealed that the mean MRT was positively associated with the herd mean 305-day milk yield and negatively associated with the feeding of grassland products in the dry period. While this study reveals valuable insights into the relationship between MRT and farm management practices, limitations such as sample size and unbalanced data distributions warrant further investigation with larger datasets and longitudinal approaches.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104382"},"PeriodicalIF":5.3,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145922947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High-throughput Verticillium wilt detection in cotton: A comparative study of faster R-CNN and YOLOv11 R-CNN和YOLOv11快速检测棉花黄萎病的比较研究

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2026-01-08 DOI: 10.1016/j.biosystemseng.2025.104379

Manish Kumar Patel , Geoff Bull , Lucy M. Egan , Natasha Swain , Vivien Rolland , Warwick N. Stiller , Warren C. Conaty

Verticillium wilt (VW), a soil-borne fungal disease of cotton, can lead to significant yield loss and has become a growing problem for global cotton production. From a global perspective, the local biotype has a high pathogenicity. Traditional phenotyping and screening methods for resistance to VW are slow, costly, and prone to human error. However, advancements in object detection models can enable automated, high-throughput screening of resistant varieties, therefore, improving speed, reducing costs, and eliminating operator bias. This study develops and evaluates the effectiveness and generalisation of two widely adopted object detection models: the two-stage Faster R-CNN and the single-stage YOLOv11 for VW in cotton stems across various backbone architectures. Digital cameras were used to collect cotton stem images from several fields. The results showed that the Faster R-CNN with the ResNet-101 model achieved a mean average precision (mAP at intersection over union (IOU) of 0.5) between 5 % and 55 % higher for the most complex YOLOv11-x and simpler YOLOv11-n, respectively, on the test dataset. Further evaluation with an independent dataset confirmed that the Faster R-CNN with ResNet-101 was the most robust and generalisable model, achieving a mAP of 85.68 %, outperforming YOLOv11 models by at least 12 % and up to 82 %. However, this enhanced mAP of the Faster R-CNN model incurred a computational cost approximately 8 % higher than that of YOLOv11-x. Nevertheless, in the context of VW detection for cotton breeding, the value of a higher mAP substantially outweighs the value of a lower computational load.

黄萎病（Verticillium wilt， VW）是一种土壤传播的棉花真菌病，可导致严重的产量损失，已成为全球棉花生产日益严重的问题。从全球范围来看，本地生物型具有较高的致病性。传统的大众抗性表型和筛选方法缓慢、昂贵，而且容易出现人为错误。然而，目标检测模型的进步可以实现抗性品种的自动化、高通量筛选，从而提高速度、降低成本并消除操作员偏见。本研究开发并评估了两种广泛采用的目标检测模型的有效性和泛化性：两阶段Faster R-CNN和单阶段YOLOv11，用于棉花茎中不同主干架构的VW。使用数码相机从几个田地收集棉花茎的图像。结果表明，对于最复杂的YOLOv11-x和最简单的YOLOv11-n，使用ResNet-101模型的更快R-CNN在测试数据集上的平均精度（mAP at intersection over union （IOU）为0.5）分别提高了5%至55%。使用独立数据集进行的进一步评估证实，使用ResNet-101的Faster R-CNN是最稳健和最通用的模型，mAP达到85.68%，比YOLOv11模型至少高出12%，最高可达82%。然而，这种增强的更快R-CNN模型的mAP比YOLOv11-x的计算成本高出约8%。然而，在棉花育种的VW检测中，较高的mAP值远远大于较低的计算负荷值。

{"title":"High-throughput Verticillium wilt detection in cotton: A comparative study of faster R-CNN and YOLOv11","authors":"Manish Kumar Patel , Geoff Bull , Lucy M. Egan , Natasha Swain , Vivien Rolland , Warwick N. Stiller , Warren C. Conaty","doi":"10.1016/j.biosystemseng.2025.104379","DOIUrl":"10.1016/j.biosystemseng.2025.104379","url":null,"abstract":"<div><div>Verticillium wilt (VW), a soil-borne fungal disease of cotton, can lead to significant yield loss and has become a growing problem for global cotton production. From a global perspective, the local biotype has a high pathogenicity. Traditional phenotyping and screening methods for resistance to VW are slow, costly, and prone to human error. However, advancements in object detection models can enable automated, high-throughput screening of resistant varieties, therefore, improving speed, reducing costs, and eliminating operator bias. This study develops and evaluates the effectiveness and generalisation of two widely adopted object detection models: the two-stage Faster R-CNN and the single-stage YOLOv11 for VW in cotton stems across various backbone architectures. Digital cameras were used to collect cotton stem images from several fields. The results showed that the Faster R-CNN with the ResNet-101 model achieved a mean average precision (mAP at intersection over union (IOU) of 0.5) between 5 % and 55 % higher for the most complex YOLOv11-x and simpler YOLOv11-n, respectively, on the test dataset. Further evaluation with an independent dataset confirmed that the Faster R-CNN with ResNet-101 was the most robust and generalisable model, achieving a mAP of 85.68 %, outperforming YOLOv11 models by at least 12 % and up to 82 %. However, this enhanced mAP of the Faster R-CNN model incurred a computational cost approximately 8 % higher than that of YOLOv11-x. Nevertheless, in the context of VW detection for cotton breeding, the value of a higher mAP substantially outweighs the value of a lower computational load.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104379"},"PeriodicalIF":5.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145922477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Good practices of artificial intelligence in biosystems engineering research 人工智能在生物系统工程研究中的良好实践

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2026-01-06 DOI: 10.1016/j.biosystemseng.2025.104373

Marta Vallejo , Chao Chen , Francisco Rovira-Más , Fernando Auat Cheein

The use and integration of Artificial Intelligence (AI) and machine learning technologies into biosystems engineering create unprecedented opportunities for modelling, optimisation, and decision support across agriculture, livestock, food systems, environmental management, and related domains. However, the increasing complexity and often opacity of these methods is raising concerns regarding scientific rigour, reproducibility, transparency, generalisation and ethical responsibility. This letter establishes a set of good practice principles for authors submitting AI-driven research to Biosystems Engineering journal. The guidelines outline essential requirements for data quality, documentation of methodologies, experimental protocols, model selection, evaluation, interpretability, and reproduction of results. They emphasise the importance of open datasets and code availability, appropriate validation strategies, meaningful novelty, and clear evidence of relevance to the Biosystems Engineering scope.

人工智能（AI）和机器学习技术在生物系统工程中的应用和集成，为农业、畜牧业、粮食系统、环境管理和相关领域的建模、优化和决策支持创造了前所未有的机会。然而，这些方法日益增加的复杂性和往往不透明引起了人们对科学严谨性、可重复性、透明度、泛化和伦理责任的担忧。这封信为作者向《生物系统工程》杂志提交人工智能驱动的研究建立了一套良好的实践原则。该指南概述了数据质量、方法文件、实验方案、模型选择、评估、可解释性和结果再现的基本要求。他们强调开放数据集和代码可用性、适当的验证策略、有意义的新颖性以及与生物系统工程范围相关的明确证据的重要性。

引用次数: 0

CFD modelling of an electrostatic spraying system to optimise pesticide spray efficiency and reduce drift 静电喷雾系统的CFD建模，以优化农药喷洒效率和减少漂移

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2026-01-02 DOI: 10.1016/j.biosystemseng.2025.104378

Matthew Herkins , Lingying Zhao , Heping Zhu , Hongyoung Jeon

Electrostatic spraying technology enhances spray efficiency and reduces airborne drift by imparting electrical charges to droplets, which increases their attraction to crop canopies. However, determining the best configuration for an electrostatic pesticide sprayer is difficult as numerous parameters impact spray efficiency. Experimental optimisation is resource-intensive and time-consuming, which makes computational modelling an excellent alternative optimisation method. This study developed a computational fluid dynamics (CFD) model to predict droplet charge-to-mass ratio (CMR), canopy deposition, and downwind drift for electrically charged sprays. The model was validated against canopy deposition and airborne drift measurement data collected in a wind tunnel at wind speeds of 0 and 2.24 m s⁻¹ using five hollow-cone nozzles and a 50 mm diameter electrode held at an applied voltage of 20 kV DC. Results showed that the model could predict the average canopy deposition from an electrostatic spraying system at specific locations within the canopy with average relative errors of 40.3 % and 58.8 % at wind speeds of 0 and 2.24 m s⁻¹, respectively. At a wind speed of 2.24 m s⁻¹, the model acceptably predicted airborne drift deposits up to a 0.70 m height, with an average relative error of 50.1 % for the validated cases; however, prediction errors increased substantially above this height. These findings demonstrate that CFD modelling is a promising method for optimising electrostatic spraying system configurations to maximise spray efficiency and minimise airborne drift, especially in low-wind environments, such as greenhouses.

静电喷雾技术提高了喷雾效率，并通过向液滴传递电荷来减少空气漂移，从而增加了液滴对作物冠层的吸引力。然而，确定静电农药喷雾器的最佳配置是困难的，因为许多参数影响喷雾效率。实验优化是一种资源密集和耗时的优化方法，这使得计算建模成为一种很好的替代优化方法。本研究开发了一个计算流体动力学（CFD）模型来预测带电喷雾的电荷质量比（CMR）、冠层沉积和顺风漂移。在风速为0和2.24 m s - 1的风洞中，使用5个空心锥喷嘴和直径为50 mm的电极，在20 kV直流电压下，对该模型进行了验证。结果表明：在风速为0 m s−1和2.24 m s−1的条件下，该模型能够预测林冠内特定位置静电喷涂系统的平均林冠沉降，平均相对误差分别为40.3%和58.8%。在风速为2.24 m s−1时，该模型可接受地预测高达0.70 m的空中漂移沉积物，验证案例的平均相对误差为50.1%；然而，在此高度以上，预测误差大大增加。这些研究结果表明，CFD建模是一种很有前途的方法，可以优化静电喷涂系统配置，以最大限度地提高喷雾效率，减少空气漂移，特别是在低风环境中，如温室。

{"title":"CFD modelling of an electrostatic spraying system to optimise pesticide spray efficiency and reduce drift","authors":"Matthew Herkins , Lingying Zhao , Heping Zhu , Hongyoung Jeon","doi":"10.1016/j.biosystemseng.2025.104378","DOIUrl":"10.1016/j.biosystemseng.2025.104378","url":null,"abstract":"<div><div>Electrostatic spraying technology enhances spray efficiency and reduces airborne drift by imparting electrical charges to droplets, which increases their attraction to crop canopies. However, determining the best configuration for an electrostatic pesticide sprayer is difficult as numerous parameters impact spray efficiency. Experimental optimisation is resource-intensive and time-consuming, which makes computational modelling an excellent alternative optimisation method. This study developed a computational fluid dynamics (CFD) model to predict droplet charge-to-mass ratio (CMR), canopy deposition, and downwind drift for electrically charged sprays. The model was validated against canopy deposition and airborne drift measurement data collected in a wind tunnel at wind speeds of 0 and 2.24 m s<sup>−1</sup> using five hollow-cone nozzles and a 50 mm diameter electrode held at an applied voltage of 20 kV DC. Results showed that the model could predict the average canopy deposition from an electrostatic spraying system at specific locations within the canopy with average relative errors of 40.3 % and 58.8 % at wind speeds of 0 and 2.24 m s<sup>−1</sup>, respectively. At a wind speed of 2.24 m s<sup>−1</sup>, the model acceptably predicted airborne drift deposits up to a 0.70 m height, with an average relative error of 50.1 % for the validated cases; however, prediction errors increased substantially above this height. These findings demonstrate that CFD modelling is a promising method for optimising electrostatic spraying system configurations to maximise spray efficiency and minimise airborne drift, especially in low-wind environments, such as greenhouses.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104378"},"PeriodicalIF":5.3,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145882464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improved and interpretable accelerometer-based farrowing prediction 改进的和可解释的基于加速度计的分娩预测

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2026-01-01 DOI: 10.1016/j.biosystemseng.2025.104381

Elisabeth Mayrhuber , Kristina Maschat , David Brunner , Stephan M. Winkler , Maciej Oczak

Predicting the onset of farrowing in sows is critical for improving animal welfare and optimising farm management. Methods driven by explainable artificial intelligence for detecting nest-building behaviour and predicting time to farrowing using accelerometer data from ear tags are presented. These methods are evaluated on a dataset containing farm management data and accelerometer data of 179 sows. During data collection the animals were kept in three different pen types with the possibility of temporary crating. By combining acceleration metrics with prepartum examinations and farm management data, a two-stage model was developed that first detects the onset of nest-building and subsequently predicted the remaining time until farrowing. Various methods, including cumulative sum (CUSUM), Bayesian estimation of abrupt change, seasonality, and trend (BEAST), and a custom model (NestDetect), were compared for nest-building detection, while symbolic regression and deep learning were used to predict farrowing time. For 82.6 % of the sows, it was possible to detect the start of nest-building behaviour in a 48-h window before the onset of farrowing. When nest-building was detected correctly, symbolic regression was able to predict the remaining time to farrowing with a mean absolute error of 9.4 h and delivered interpretable results, while NNs achieved a mean absolute error of 9.6 h without being inherently interpretable. This work emphasises the importance of model interpretability and explainability in precision livestock farming, highlighting that transparent models can facilitate timely, data-driven interventions, while having the same prediction power as non-interpretable models.

预测母猪分娩的开始对改善动物福利和优化农场管理至关重要。提出了由可解释的人工智能驱动的方法，用于检测筑巢行为和使用耳标签的加速度计数据预测分娩时间。这些方法在包含179头母猪的农场管理数据和加速度计数据的数据集上进行了评估。在数据收集期间，动物被关在三种不同类型的围栏中，可能是临时的板条箱。通过将加速指标与准备检查和农场管理数据相结合，开发了一个两阶段模型，首先检测筑巢的开始，然后预测到分娩的剩余时间。比较了各种方法，包括累积和（CUSUM）、突变、季节性和趋势的贝叶斯估计（BEAST）和自定义模型（NestDetect），用于筑巢检测，而符号回归和深度学习用于预测产仔时间。对于82.6%的母猪，可以在分娩开始前48小时的窗口内检测到筑巢行为的开始。当正确检测到筑巢时，符号回归能够预测剩余的分娩时间，平均绝对误差为9.4小时，并且提供了可解释的结果，而神经网络的平均绝对误差为9.6小时，但本质上是不可解释的。这项工作强调了模型可解释性和可解释性在精准畜牧业中的重要性，强调透明模型可以促进及时的、数据驱动的干预，同时与不可解释性模型具有相同的预测能力。

{"title":"Improved and interpretable accelerometer-based farrowing prediction","authors":"Elisabeth Mayrhuber , Kristina Maschat , David Brunner , Stephan M. Winkler , Maciej Oczak","doi":"10.1016/j.biosystemseng.2025.104381","DOIUrl":"10.1016/j.biosystemseng.2025.104381","url":null,"abstract":"<div><div>Predicting the onset of farrowing in sows is critical for improving animal welfare and optimising farm management. Methods driven by explainable artificial intelligence for detecting nest-building behaviour and predicting time to farrowing using accelerometer data from ear tags are presented. These methods are evaluated on a dataset containing farm management data and accelerometer data of 179 sows. During data collection the animals were kept in three different pen types with the possibility of temporary crating. By combining acceleration metrics with prepartum examinations and farm management data, a two-stage model was developed that first detects the onset of nest-building and subsequently predicted the remaining time until farrowing. Various methods, including cumulative sum (CUSUM), Bayesian estimation of abrupt change, seasonality, and trend (BEAST), and a custom model (NestDetect), were compared for nest-building detection, while symbolic regression and deep learning were used to predict farrowing time. For 82.6 % of the sows, it was possible to detect the start of nest-building behaviour in a 48-h window before the onset of farrowing. When nest-building was detected correctly, symbolic regression was able to predict the remaining time to farrowing with a mean absolute error of 9.4 h and delivered interpretable results, while NNs achieved a mean absolute error of 9.6 h without being inherently interpretable. This work emphasises the importance of model interpretability and explainability in precision livestock farming, highlighting that transparent models can facilitate timely, data-driven interventions, while having the same prediction power as non-interpretable models.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104381"},"PeriodicalIF":5.3,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145882441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

From segmentation to classification: Morphological phenotype extraction and classification analysis of tiny poplar seeds using the MP-Seed segmentation algorithm 从分割到分类：利用MP-Seed分割算法提取杨树微小种子的形态表型及分类分析

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2025-12-31 DOI: 10.1016/j.biosystemseng.2025.104376

Zanpeng Li, Mengmeng Qiao, Xiwei Wang, Mubikayi Muhong Horly, Maocheng Zhao, Bin Wu

Extracting poplar seed morphological phenotypes is a core task in modern poplar breeding research. Accurate seed image segmentation is crucial for phenotype extraction and data quality. However, the small size of poplar seeds and their tendency to form dense clusters challenge the accuracy of current segmentation methods. Unlike current approaches that struggle with small-target segmentation and boundary delineation, this study develops the MP-Seed semantic segmentation algorithm, which combines a small-target attention module (based on Layer Across Feature Map Attention) with a multi-task learning mechanism that integrates boundary features. This novel integration targets small-seed key regions, fuses boundary features, and refines predictions to precisely segment densely clustered seeds, achieving superior accuracy and fine-grained delineation compared to current single-task methods. To address low efficiency and accuracy in poplar seed morphological phenotype extraction, this study further proposes a high-throughput extraction method leveraging the MP-Seed algorithm. To analyse the phenotypic data, an SVM classification model classifies eight types of poplar seeds. Experimental validation shows that the MP-Seed algorithm outperforms current methods on the test set, achieving Seed_IoU of 94.1 %, mIoU of 97.2 %, and Reference_IoU of 97.6 %. The high-throughput phenotyping method measures seed length and width with relative errors within 2.72 % versus manual measurements and extracts ten morphological traits at about 18.3 seeds per second. The overall classification accuracy reaches 91.1 %. Overall, this study provides technical support for accurate poplar seed segmentation and efficient morphological phenotype extraction, offering a valuable reference for other seed morphological phenotype research and analysis.

杨树种子形态表型提取是现代杨树育种研究的核心内容。准确的种子图像分割对表型提取和数据质量至关重要。然而，杨树种子的体积小，易于形成密集的簇，这对目前的分割方法的准确性提出了挑战。与目前的小目标分割和边界划分方法不同，本研究开发了MP-Seed语义分割算法，该算法将小目标注意模块（基于跨层特征映射注意）与集成边界特征的多任务学习机制相结合。这种新颖的集成针对小种子关键区域，融合边界特征，并细化预测，以精确分割密集聚集的种子，与当前的单任务方法相比，实现了更高的准确性和细粒度描述。针对杨树种子形态表型提取效率低、准确性低的问题，本研究进一步提出了一种利用MP-Seed算法的高通量提取方法。为了分析表型数据，利用SVM分类模型对8种杨树种子进行了分类。实验验证表明，MP-Seed算法在测试集上优于现有方法，Seed_IoU为94.1%，mIoU为97.2%，Reference_IoU为97.6%。高通量表型方法测量种子长度和宽度，相对于人工测量误差在2.72%以内，提取10个形态性状的速度约为每秒18.3颗种子。总体分类准确率达到91.1%。总体而言，本研究为杨树种子的准确切分和高效形态表型提取提供了技术支持，为其他种子形态表型研究和分析提供了有价值的参考。

{"title":"From segmentation to classification: Morphological phenotype extraction and classification analysis of tiny poplar seeds using the MP-Seed segmentation algorithm","authors":"Zanpeng Li, Mengmeng Qiao, Xiwei Wang, Mubikayi Muhong Horly, Maocheng Zhao, Bin Wu","doi":"10.1016/j.biosystemseng.2025.104376","DOIUrl":"10.1016/j.biosystemseng.2025.104376","url":null,"abstract":"<div><div>Extracting poplar seed morphological phenotypes is a core task in modern poplar breeding research. Accurate seed image segmentation is crucial for phenotype extraction and data quality. However, the small size of poplar seeds and their tendency to form dense clusters challenge the accuracy of current segmentation methods. Unlike current approaches that struggle with small-target segmentation and boundary delineation, this study develops the MP-Seed semantic segmentation algorithm, which combines a small-target attention module (based on Layer Across Feature Map Attention) with a multi-task learning mechanism that integrates boundary features. This novel integration targets small-seed key regions, fuses boundary features, and refines predictions to precisely segment densely clustered seeds, achieving superior accuracy and fine-grained delineation compared to current single-task methods. To address low efficiency and accuracy in poplar seed morphological phenotype extraction, this study further proposes a high-throughput extraction method leveraging the MP-Seed algorithm. To analyse the phenotypic data, an SVM classification model classifies eight types of poplar seeds. Experimental validation shows that the MP-Seed algorithm outperforms current methods on the test set, achieving Seed_IoU of 94.1 %, mIoU of 97.2 %, and Reference_IoU of 97.6 %. The high-throughput phenotyping method measures seed length and width with relative errors within 2.72 % versus manual measurements and extracts ten morphological traits at about 18.3 seeds per second. The overall classification accuracy reaches 91.1 %. Overall, this study provides technical support for accurate poplar seed segmentation and efficient morphological phenotype extraction, offering a valuable reference for other seed morphological phenotype research and analysis.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104376"},"PeriodicalIF":5.3,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145882440","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DEM modelling of crop residue and soil dynamics as affected by the tillage direction of a disc harrow 盘耙耕作方向对作物残茬和土壤动力学影响的DEM模拟

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2025-12-30 DOI: 10.1016/j.biosystemseng.2025.104369

Peng Wu , Xirui Zhang , Ying Chen

A comprehensive understanding of crop residue and soil dynamics under tillage is essential for improving the operational efficiency of tillage implements. In this study, a tandem disc harrow and its interaction with wheat-residue in a sandy loam soil were modelled using the discrete element method. The model was able to simulate two tillage directions: perpendicular and parallel to standing wheat-stubble rows (Perp-Direction and Para-Direction). The model was validated using field measured data. Field experiments showed that Perp-Direction produced significantly lower residue cover than Para-Direction (P < 0.05). Both tillage directions resulted in similar stubble forward displacements (mean: 354 mm). The DEM model predicted these variables with an overall relative error of 19.7 %. Simulations further revealed that stubble trajectories and soil cutting forces varied among individual discs due to the tandem configuration of the harrow. These variables were also influenced by operational parameters: disc angle, harrow travelling speed, and working depth. Reducing these operational parameters decreased soil surface roughness in both tillage directions. Conversely, increasing these operational parameters required higher total draft forces and disturbed a larger soil area. However, the soil cutting efficiency, a performance index combining both total draft force and soil disturbance area, declined from 95.9 to 37.3 m³ MJ⁻¹ under these conditions. Overall, these findings provide useful guidance for optimising tillage direction and operational parameters of tandem disc harrows to enhance field performance.

全面了解耕作条件下的作物残茬和土壤动态对提高耕作工具的使用效率至关重要。本文采用离散元法对沙地壤土中串联盘耙及其与小麦残茬的相互作用进行了数值模拟。该模型能够模拟两种耕作方向：垂直和平行于直立小麦茬行（垂直方向和平行方向）。利用现场实测数据对模型进行了验证。田间试验表明，Perp-Direction的残茬盖度显著低于Para-Direction （P < 0.05）。两种耕作方式产生的残茬向前位移相似（平均值：354mm）。DEM模型预测这些变量的总体相对误差为19.7%。模拟进一步表明，由于耙的串联配置，个别圆盘的割茬轨迹和土壤切割力有所不同。这些变量还受到操作参数的影响：圆盘角度、耙移动速度和工作深度。减少这些操作参数降低了两个耕作方向的土壤表面粗糙度。相反，增加这些操作参数需要更高的总牵引力，并扰动更大的土壤面积。而综合总风力和土壤扰动面积的土壤切削效率从95.9 m3 MJ−1下降到37.3 m3 MJ−1。综上所述，研究结果可为优化连作盘耙的耕作方向和操作参数，提高田间性能提供有益的指导。

{"title":"DEM modelling of crop residue and soil dynamics as affected by the tillage direction of a disc harrow","authors":"Peng Wu , Xirui Zhang , Ying Chen","doi":"10.1016/j.biosystemseng.2025.104369","DOIUrl":"10.1016/j.biosystemseng.2025.104369","url":null,"abstract":"<div><div>A comprehensive understanding of crop residue and soil dynamics under tillage is essential for improving the operational efficiency of tillage implements. In this study, a tandem disc harrow and its interaction with wheat-residue in a sandy loam soil were modelled using the discrete element method. The model was able to simulate two tillage directions: perpendicular and parallel to standing wheat-stubble rows (Perp-Direction and Para-Direction). The model was validated using field measured data. Field experiments showed that Perp-Direction produced significantly lower residue cover than Para-Direction (P < 0.05). Both tillage directions resulted in similar stubble forward displacements (mean: 354 mm). The DEM model predicted these variables with an overall relative error of 19.7 %. Simulations further revealed that stubble trajectories and soil cutting forces varied among individual discs due to the tandem configuration of the harrow. These variables were also influenced by operational parameters: disc angle, harrow travelling speed, and working depth. Reducing these operational parameters decreased soil surface roughness in both tillage directions. Conversely, increasing these operational parameters required higher total draft forces and disturbed a larger soil area. However, the soil cutting efficiency, a performance index combining both total draft force and soil disturbance area, declined from 95.9 to 37.3 m<sup>3</sup> MJ<sup>−1</sup> under these conditions. Overall, these findings provide useful guidance for optimising tillage direction and operational parameters of tandem disc harrows to enhance field performance.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104369"},"PeriodicalIF":5.3,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145882465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Influence of film-tensioning lines on the wind-induced responses of flat-elliptical pipe greenhouse 膜张拉线对扁平椭圆管温室风致响应的影响

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2025-12-30 DOI: 10.1016/j.biosystemseng.2025.104370

Hengyan Xie , Cunxing Wei , Xin Zheng , Wenbao Xu

Under extreme wind loads, plastic greenhouses frequently experience film uplift, which leads to structural instability and crop damage. Existing research predominantly focuses on the stability of the greenhouse skeletons and the effect of covering material on load-bearing capacity, while the significant role of the film-tensioning lines in enhancing the wind resistance of plastic greenhouses has received limited attention. This study investigates the wind-induced response of plastic greenhouses by utilising ABAQUS finite element software to establish two models: a film-skeletons (FS) model and a film-tensioning lines-film-skeletons (FFS) model. Both static and dynamic wind load analyses are performed based on the Davenport wind spectrum, using a linear filtering method to simulate fluctuating wind speeds. The study compares the wind-induced responses of various components and analyses the contact conditions between the film and the skeletons under different loading scenarios. The results demonstrate that the introduction of the film-tensioning lines significantly enhances the film's stiffness, optimises the stress distribution, and effectively suppresses excessive deformation of the film. Additionally, the film-tensioning lines alleviates stress concentration in the skeletons, limits skeletons displacement, increases the radial constraint on the film, and reduces shear stress between the film and the skeletons. Under dynamic wind loads, the reinforcing effect of the film-tensioning lines on both film and skeletons stiffness is even more pronounced. This research contributes to the theoretical skeletons for analysing the wind-induced responses of greenhouse components, providing a scientific basis for the accurate evaluation of the wind resistance of plastic greenhouses.

在极端风荷载作用下，塑料大棚经常发生薄膜隆起，导致结构失稳和作物受损。现有的研究主要集中在温室骨架的稳定性和覆盖材料对承载能力的影响上，而膜张拉线在提高塑料大棚抗风能力方面的重要作用却很少得到关注。本研究利用ABAQUS有限元软件对塑料大棚的风致响应进行了研究，建立了膜-骨架（FS）模型和膜-张拉线-膜-骨架（FFS）模型。基于达文波特风谱，采用线性滤波方法模拟脉动风速，进行了静态和动态风荷载分析。比较了不同荷载下各构件的风致响应，分析了膜层与骨架的接触条件。结果表明，膜张紧线的引入显著提高了膜的刚度，优化了膜的应力分布，有效地抑制了膜的过度变形。此外，膜张紧线缓解了骨架中的应力集中，限制了骨架的位移，增加了膜的径向约束，减小了膜与骨架之间的剪应力。在动风荷载作用下，膜张拉线对膜和骨架刚度的增强作用更为明显。本研究为分析温室构件的风致响应提供了理论框架，为准确评价塑料大棚的抗风能力提供了科学依据。

{"title":"Influence of film-tensioning lines on the wind-induced responses of flat-elliptical pipe greenhouse","authors":"Hengyan Xie , Cunxing Wei , Xin Zheng , Wenbao Xu","doi":"10.1016/j.biosystemseng.2025.104370","DOIUrl":"10.1016/j.biosystemseng.2025.104370","url":null,"abstract":"<div><div>Under extreme wind loads, plastic greenhouses frequently experience film uplift, which leads to structural instability and crop damage. Existing research predominantly focuses on the stability of the greenhouse skeletons and the effect of covering material on load-bearing capacity, while the significant role of the film-tensioning lines in enhancing the wind resistance of plastic greenhouses has received limited attention. This study investigates the wind-induced response of plastic greenhouses by utilising ABAQUS finite element software to establish two models: a film-skeletons (FS) model and a film-tensioning lines-film-skeletons (FFS) model. Both static and dynamic wind load analyses are performed based on the Davenport wind spectrum, using a linear filtering method to simulate fluctuating wind speeds. The study compares the wind-induced responses of various components and analyses the contact conditions between the film and the skeletons under different loading scenarios. The results demonstrate that the introduction of the film-tensioning lines significantly enhances the film's stiffness, optimises the stress distribution, and effectively suppresses excessive deformation of the film. Additionally, the film-tensioning lines alleviates stress concentration in the skeletons, limits skeletons displacement, increases the radial constraint on the film, and reduces shear stress between the film and the skeletons. Under dynamic wind loads, the reinforcing effect of the film-tensioning lines on both film and skeletons stiffness is even more pronounced. This research contributes to the theoretical skeletons for analysing the wind-induced responses of greenhouse components, providing a scientific basis for the accurate evaluation of the wind resistance of plastic greenhouses.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104370"},"PeriodicalIF":5.3,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145882442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Quantitative assessment and predictive modelling of stem damage during seedling separation in mechanical rice transplanting 水稻机械插秧分苗过程中茎损伤的定量评估与预测模型

IF 5.3 1区农林科学 Q1 AGRICULTURAL ENGINEERING

Biosystems Engineering

Pub Date : 2025-12-26 DOI: 10.1016/j.biosystemseng.2025.104377

Dongdong Xi, Jinnan Que, Ruiling Shen, Fuming Kuang, Wei Xiong, Shun Zhang, Dequan Zhu

Rice seedling stems are particularly vulnerable to structural damage during the seedling separation phase of mechanical transplanting, especially under non-ideal plant-machine interactions. Owing to its internal and transient nature, such damage is inherently difficult to quantify or predict. This study presents a novel modelling framework for stem damage assessment, which establishes a quantitative relationship between the maximum impact load (F_max) during seedling separation and internal damage severity, quantified by the damaged area ratio (D_ar). High-speed imaging and triaxial force sensors were employed to measure F_max across seedlings aged 20, 30 and 40 d under varying transplanting speeds. Microscopic cross-sections of stems were analysed to calculate D_ar. A composite impact force model, incorporating stem bending rigidity, lateral needle–stem offset and contact duration, was developed to support experimental design. A strong positive correlation was observed between F_max and D_ar across all seedling age groups (ρ > 0.93, p < 0.001). At 100–150 rpm, D_ar generally remained below 2 %, whereas at higher speeds damage increased and at 300 rpm, 40 d seedlings showed a median D_ar > 8 %. Age-specific linear regression models achieved high predictive accuracy and good calibration (cross-validated R² of 0.86–0.91; RMSE of 0.33–0.73 percentage points in D_ar), while extending these models with a restricted cubic spline further reduced errors in the upper damage tail. This framework offers theoretical insights into age- and speed-dependent stem damage and practical tools for optimising transplanting parameters and supporting real-time, damage-aware control strategies to mitigate mechanical damage risk and improve seedling survival and post-transplant performance.

在机械插秧分苗阶段，特别是在非理想的机苗相互作用下，水稻幼苗茎秆特别容易受到结构损伤。由于其内部和瞬态性质，这种损害本质上是难以量化或预测的。本研究提出了一种新的茎损伤评估模型框架，该框架建立了幼苗分离时最大冲击载荷（Fmax）与内部损伤严重程度之间的定量关系，并用损伤面积比（Dar）量化。采用高速成像和三轴力传感器测量不同移栽速度下幼苗20d、30d和40d的Fmax。通过分析茎的微观截面来计算Dar。为支持实验设计，建立了包含杆弯曲刚度、横向针杆偏移量和接触时间的复合冲击力模型。Fmax与Dar在各苗龄组间呈显著正相关（ρ > 0.93, p < 0.001）。在100-150转/分时，Dar一般保持在2%以下，而在更高的转速下，损伤增加，在300转/分时，40 d幼苗的Dar中值为8%。特定年龄的线性回归模型具有较高的预测精度和良好的校准（交叉验证的R2为0.86-0.91；在Dar中RMSE为0.33-0.73个百分点），而用受限三次样条扩展这些模型进一步减少了上损伤尾的误差。该框架为年龄和速度相关的茎损伤提供了理论见解，并为优化移栽参数和支持实时、损伤感知控制策略提供了实用工具，以减轻机械损伤风险，提高幼苗存活率和移栽后性能。

{"title":"Quantitative assessment and predictive modelling of stem damage during seedling separation in mechanical rice transplanting","authors":"Dongdong Xi, Jinnan Que, Ruiling Shen, Fuming Kuang, Wei Xiong, Shun Zhang, Dequan Zhu","doi":"10.1016/j.biosystemseng.2025.104377","DOIUrl":"10.1016/j.biosystemseng.2025.104377","url":null,"abstract":"<div><div>Rice seedling stems are particularly vulnerable to structural damage during the seedling separation phase of mechanical transplanting, especially under non-ideal plant-machine interactions. Owing to its internal and transient nature, such damage is inherently difficult to quantify or predict. This study presents a novel modelling framework for stem damage assessment, which establishes a quantitative relationship between the maximum impact load (<em>F</em><sub>max</sub>) during seedling separation and internal damage severity, quantified by the damaged area ratio (<em>D</em><sub>ar</sub>). High-speed imaging and triaxial force sensors were employed to measure <em>F</em><sub>max</sub> across seedlings aged 20, 30 and 40 d under varying transplanting speeds. Microscopic cross-sections of stems were analysed to calculate <em>D</em><sub>ar</sub>. A composite impact force model, incorporating stem bending rigidity, lateral needle–stem offset and contact duration, was developed to support experimental design. A strong positive correlation was observed between <em>F</em><sub>max</sub> and <em>D</em><sub>ar</sub> across all seedling age groups (<em>ρ</em> > 0.93, <em>p</em> < 0.001). At 100–150 rpm, <em>D</em><sub>ar</sub> generally remained below 2 %, whereas at higher speeds damage increased and at 300 rpm, 40 d seedlings showed a median <em>D</em><sub>ar</sub> > 8 %. Age-specific linear regression models achieved high predictive accuracy and good calibration (cross-validated <em>R</em><sup>2</sup> of 0.86–0.91; RMSE of 0.33–0.73 percentage points in <em>D</em><sub>ar</sub>), while extending these models with a restricted cubic spline further reduced errors in the upper damage tail. This framework offers theoretical insights into age- and speed-dependent stem damage and practical tools for optimising transplanting parameters and supporting real-time, damage-aware control strategies to mitigate mechanical damage risk and improve seedling survival and post-transplant performance.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"263 ","pages":"Article 104377"},"PeriodicalIF":5.3,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0